Rice phenological period growth vigor monitoring system and method based on unmanned aerial vehicle

Abstract

The invention relates to the field of rice growth vigor monitoring, and discloses a rice phenological period growth vigor monitoring system and method based on an unmanned aerial vehicle, and the system comprises a rice data collection module which divides a rice field into a plurality of subareas, and carries out the image collection of each subarea through a sensor carried by the unmanned aerial vehicle; the rice image processing module is used for transmitting the collected image to a ground station and storing the collected image, and performing preprocessing operations of cutting, size adjustment and edge cutting on the collected image by using an image processing tool to obtain an image in a standard format; the rice data analysis module analyzes the preprocessed standard image, calculates a crop evaluation coefficient of each partition, and evaluates the growth vigor of the crops according to the crop evaluation coefficients; and the coping strategy module is used for judging whether the rice in the current subarea has the risk of agricultural disasters or not according to the evaluation result, and performing corresponding fertilization, irrigation or pest control according to the types of the agricultural disasters.

Description

Technical Field

[0001] The present invention relates to the field of rice growth monitoring, and in particular to a rice phenological growth monitoring system and method based on an unmanned aerial vehicle. Background Art

[0002] Precision agriculture is a new trend in the development of modern agriculture. It uses modern information technology and intelligent equipment to monitor and manage the growth of large-scale food crops, such as rice, in real time, accurately and quickly.

[0003] Traditional crop growth monitoring methods rely on professionals to go to the fields for visual identification. This method is inefficient, costly, and has large errors when obtaining rice growth information over a large area and across multiple varieties. In addition, traditional vegetation physiological and biochemical parameter extraction methods often cause irreparable damage to vegetation and cannot effectively reflect real-time vegetation information.

[0004] With the development of remote sensing technology, satellite remote sensing has become an important means of agricultural monitoring and evaluation. However, due to the limitation of resolution, traditional satellite remote sensing is insufficient in the detailed description and real-time data acquisition of small-scale crops such as fields, plots, and plants.

[0005] As a new technical means, UAV remote sensing technology has shown great advantages in farmland monitoring with its low cost, strong timeliness and high resolution. UAVs can be equipped with a variety of sensors such as digital cameras, multispectral cameras, and hyperspectral cameras to provide high-precision and high-resolution data support for the collection of agricultural information. However, in actual application, there is a situation where only single image data obtained by UAVs is used, while ignoring the actual soil environment and pests and diseases, which ultimately leads to inaccurate assessment of rice growth, which in turn affects the subsequent corresponding management measures based on rice growth. Summary of the invention

[0006] The purpose of the present invention is to provide a rice phenological growth monitoring system and method based on unmanned aerial vehicle to solve the above technical problems.

[0007] The purpose of the present invention can be achieved by the following technical solutions:

[0008] A rice phenological growth monitoring system based on drone, comprising:

[0009] The rice data collection module divides the rice field into several zones, and uses drones to fly low over the rice field according to a preset flight route and altitude, while the sensors carried by the drones collect images of each zone;

[0010] The rice image processing module is used to transmit the collected images to the ground station and store them, and use image processing tools to perform pre-processing operations such as cropping, resizing, and edge cutting on the collected images to obtain images in a standard format;

[0011] The rice data analysis module is used to analyze the pre-processed standard image and calculate the crop evaluation coefficient of each partition, and evaluate the growth of the crop according to the crop evaluation coefficient;

[0012] The response strategy module determines whether the rice in the current partition is at risk of agricultural disasters based on the assessment results; if there is a risk of agricultural disasters, corresponding fertilization, irrigation or pest and disease control is carried out according to the type of agricultural disaster, which includes lodging, drought and insect pests.

[0013] As a further technical solution, the process of obtaining the crop evaluation coefficient for each partition is:

[0014] By formula:

[0015]

[0016] Calculate the crop assessment coefficient Q for each partition i ;

[0017] Among them, i is the i-th partition, VC i is the vegetation coverage index for each partition, α i is the level of rice growth potential in each partition; when the rice growth potential level is low, α i =1, when the rice growth potential level is medium, α i =2, when the rice growth potential level is high, α i =3;

[0018] VC ith is the standard value of vegetation coverage index, DE th DE is the reference value of soil environmental degradation index. i is the soil environmental degradation index for each sub-area, FH ik For each sub-area, the pest risk index is is the average value of the pest risk index, and μ and ρ are adjustment coefficients.

[0019] As a further technical solution, the process of determining the rice growth potential level in each partition is as follows:

[0020] The processed image was binarized using the adaptive threshold method Otsu to obtain a binary image, in which white pixels represent the rice area and black pixels represent the background area;

[0021] According to the pixel value R of the white area in the binary imagej Calculate the vegetation coverage index VC for each partition i , the expression is: Among them, W i is the total pixel value, n is the number of images, and j is the jth image;

[0022] The vegetation coverage index VC of each partition i With the preset preparation coverage threshold interval [VC min , VC max ] for comparison;

[0023] If VC i >VC max , then the rice growth potential is judged to be advanced;

[0024] If VC i ∈[VC min , VC max ], the rice growth potential is judged to be medium;

[0025] If VC i <VC min , then the level of rice growth potential is judged to be low.

[0026] As a further technical solution, the process of evaluating the growth of crops according to the crop evaluation coefficient is as follows:

[0027] The crop evaluation coefficient Q for each partition obtained will be calculated i With the preset crop assessment threshold Q 0 Make comparisons;

[0028] If Q i ≥Q 0 , it is estimated that the crops in the current partition are growing well;

[0029] If Q i 0 , then a preliminary assessment is made on the poor growth of crops in the current sub-area.

[0030] As a further technical solution, based on the assessment results, the process of determining whether the rice in the current zone has an agricultural disaster risk is as follows:

[0031] A secondary judgment is made on the zones where the crop growth is poor. The specific process is as follows:

[0032] The vegetation coverage index VC of each partition i , soil environmental degradation index DE i , Pest risk index FH ik The corresponding warning values ​​VC 0 ,DE 0 , FH​0 Make comparisons;

[0033] If VC i <VC 0 And DE i >DE 0 And FH i ≤FH 0 , then it is judged that the rice in the current zone is at risk of drought;

[0034] If VC i <VC 0 And DE i ≤DE 0 And FH i >FH 0 , then it is judged that the rice in the current partition has the risk of pests and diseases;

[0035] If VC i ≥VC 0 And DE i ≤DE 0 And FH i ≤FH 0 , it is judged that the rice in the current partition is at risk of lodging;

[0036] If VC i <VC 0 And DE i >DE 0 And FH i >FH 0 , an emergency alarm is immediately issued to the current partition.

[0037] As a further technical solution, the process of obtaining the soil environmental degradation index for each partition is as follows:

[0038] Fitting to obtain the variation curve of soil moisture parameters over time for each partition B ω (t), and substitute it into the following formula:

[0039]

[0040] Calculate the soil environmental degradation index DE for each partition i ;

[0041] Among them, t0~t1 is the detection period, and Δθ is the reference value of the soil moisture parameter change between t0 and t1 determined based on historical data analysis.

[0042] As a further technical solution, obtain the pest risk index FH for each partition i The process is:

[0043] Use the pre-trained convolutional neural network model to identify pests and diseases in standard format images, obtain the location information of each pest and disease, and then count the number of pests and diseases K1;

[0044] By formula:

[0045]

[0046] Among them, γ is the correction coefficient, N is the number of sampling times, s is the sth sampling, de is the density of pests and diseases, K1 is the number of pests and diseases in the current image, AZ is the area of ​​the current image, is the mean density of pests and diseases.

[0047] A method for monitoring rice phenological growth based on an unmanned aerial vehicle comprises the following steps:

[0048] S1. Divide the rice field into several zones, and use the drone to fly low over the rice field according to the preset flight route and altitude, while the onboard sensor collects images of each zone;

[0049] S2, transmitting the collected images to the ground station and storing them, using image processing tools to perform pre-processing operations such as cropping, resizing, and edge cutting on the collected images to obtain images in a standard format;

[0050] S3, analyzing the preprocessed standard image, calculating the crop evaluation coefficient of each partition, and evaluating the growth of the crop according to the crop evaluation coefficient;

[0051] S4. Based on the assessment results, determine whether the rice in the current sub-area is at risk of agricultural disasters;

[0052] If there is a risk of agricultural disasters, appropriate fertilization, irrigation or pest and disease control will be carried out according to the type of agricultural disaster.

[0053] Beneficial effects of the present invention:

[0054] The present invention realizes efficient and comprehensive monitoring of rice fields by carrying sensors on unmanned aerial vehicles for low-altitude flight to collect images. Compared with traditional manual monitoring methods, it can improve monitoring efficiency, reduce labor costs, and make monitoring work more convenient and efficient; and the collected images are preprocessed by the rice image processing module to obtain images in a standard format, which provides a reliable basis for subsequent data analysis; at the same time, the preprocessed standard images are deeply analyzed by the rice data analysis module, and the crop evaluation coefficient of each partition is calculated, so as to accurately evaluate the growth of the crops; finally, the response strategy module can quickly determine whether the rice in the current partition is at risk of agricultural disasters based on the evaluation results, and formulate corresponding response strategies according to the type of disaster, so as to minimize the impact of agricultural disasters on rice growth and ensure the yield and quality of rice. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] The present invention will be further described below in conjunction with the accompanying drawings.

[0056] Figure 1 is a system block diagram of the present invention;

[0057] Figure 2 It is a diagram of the method steps of the present invention. DETAILED DESCRIPTION

[0058] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0059] See also Figure 1-Figure 2 As shown, the present invention is a rice phenological growth monitoring system based on drone, comprising:

[0060] The rice data collection module divides the rice field into several zones, and uses drones to fly low over the rice field according to a preset flight route and altitude, while the sensors carried by the drones collect images of each zone;

[0061] The rice image processing module is used to transmit the collected images to the ground station and store them. The collected images are pre-processed by cropping, resizing, and cutting off the edges using image processing tools such as OpenCV and Pillow to obtain images in a standard format.

[0062] The rice data analysis module is used to analyze the pre-processed standard image and calculate the crop evaluation coefficient of each partition, and evaluate the growth of the crop according to the crop evaluation coefficient;

[0063] The response strategy module determines whether the rice in the current partition is at risk of agricultural disasters based on the assessment results; if there is a risk of agricultural disasters, corresponding fertilization, irrigation or pest and disease control is carried out according to the type of agricultural disaster, which includes lodging, drought and insect pests.

[0064] In this embodiment, the image collection by low-altitude flight of unmanned aerial vehicles equipped with sensors is realized, so that efficient and comprehensive monitoring of rice fields is achieved. Compared with the traditional manual monitoring method, it can improve the monitoring efficiency, reduce the labor cost, and make the monitoring work more convenient and efficient; and the collected images are preprocessed by the rice image processing module to obtain images in a standard format, which provides a reliable basis for subsequent data analysis; at the same time, the preprocessed standard images are deeply analyzed by the rice data analysis module, and the crop evaluation coefficient of each partition is calculated, so as to accurately evaluate the growth of the crops; finally, the response strategy module can quickly determine whether the rice in the current partition is at risk of agricultural disasters based on the evaluation results, and formulate corresponding response strategies according to the type of disaster, so as to minimize the impact of agricultural disasters on rice growth and ensure the yield and quality of rice.

[0065] The process of obtaining the crop evaluation coefficient for each partition is:

[0066] By formula:

[0067]

[0068] Calculate the crop assessment coefficient Q for each partition i ;

[0069] Among them, i is the i-th partition, VC i is the vegetation coverage index for each partition, α i is the level of rice growth potential in each partition; when the rice growth potential level is low, α i =1, when the rice growth potential level is medium, α i =2, when the rice growth potential level is high, α i =3;

[0070] VC ith is the standard value of vegetation coverage index, DE th DE is the reference value of soil environmental degradation index. i is the soil environmental degradation index for each sub-area, FH ik For each sub-area, the pest risk index is is the average value of the pest risk index, μ and ρ are adjustment coefficients, which are comprehensively formulated based on historical data and experimental data. The initial value can be set to μ=ρ=1.

[0071] In this embodiment, the parameters of soil environment, vegetation coverage and pests are unified to comprehensively determine the crop evaluation coefficient for each partition, so as to achieve an accurate evaluation of the growth of crops. Specifically, by calculating Get the crop assessment coefficient Q for each partition i ,use Get the actual vegetation coverage index in each partition and the corresponding rice growth potential level in each partition. The higher the level, the higher the α i The larger the value, the faster the rice grows in the sub-area; the larger the soil environment degradation index is, the closer it is to the reference value of the soil environment degradation index, which means that the soil environment in the sub-area is poor and the probability of affecting crop production is high; reuse Calculate the fluctuation of pests and diseases in each sub-area. Obviously, the greater the fluctuation, the greater the possibility of a sudden increase in pests and diseases, and the greater the impact on rice growth.

[0072] The process of determining the rice growth potential level in each partition is as follows:

[0073] The processed image was binarized using the adaptive threshold method Otsu to obtain a binary image, in which white pixels represent the rice area and black pixels represent the background area;

[0074] According to the pixel value R of the white area in the binary image j Calculate the vegetation coverage index VC for each partition i , the expression is: Among them, W i is the total pixel value, n is the number of images, and j is the jth image;

[0075] The vegetation coverage index VC of each partition i With the preset preparation coverage threshold interval [VC min , VC max ] for comparison;

[0076] If VC i >VC max , then the rice growth potential is judged to be advanced;

[0077] If VC i ∈[VC min , VC max ], the rice growth potential is judged to be medium;

[0078] If VC i <VC min , then the level of rice growth potential is judged to be low.

[0079] In this embodiment, the adaptive threshold method Otsu can automatically determine the optimal threshold according to the grayscale histogram of the image, thereby realizing the binarization of the image, thereby reducing the subjectivity of manually setting the threshold and improving the accuracy of the binary image; at the same time, the vegetation coverage rate is calculated based on the pixel value of the white area in the binary image, which can accurately reflect the growth of rice and provide a reliable basis for the subsequent growth potential judgment; and then Calculate the vegetation coverage index VC of each partition i , and then prepare the coverage threshold interval [VC min , VC max ] comparison results can realize rapid judgment of rice growth potential level and improve efficiency.

[0080] The process of evaluating the growth of crops according to the crop evaluation coefficient is as follows:

[0081] The crop evaluation coefficient Q for each partition obtained will be calculated i With the preset crop assessment threshold Q 0 Make comparisons;

[0082] If Q i ≥Q 0 , it is estimated that the crops in the current partition are growing well;

[0083] If Q i 0 , then a preliminary assessment is made on the poor growth of crops in the current sub-area.

[0084] Through the above technical solution, the rapid judgment and division of crop growth in each zone is achieved, thus providing effective support for the subsequent selection of agricultural disaster strategies.

[0085] According to the assessment results, the process of judging whether there is a risk of agricultural disasters in the current sub-area is as follows:

[0086] A secondary judgment is made on the zones where the crop growth is poor. The specific process is as follows:

[0087] The vegetation coverage index VC of each partition i , soil environmental degradation index DE i , Pest risk index FH ik The corresponding warning values ​​VC 0 ,DE 0 , FH 0 Make comparisons;

[0088] If VC i <VC 0 And DE i >DE 0 And FH i ​≤FH 0 , then it is judged that the rice in the current zone is at risk of drought;

[0089] If VC i <VC 0 And DE i ≤DE 0 And FH i >FH 0 , then it is judged that the rice in the current partition has the risk of pests and diseases;

[0090] If VC i ≥VC 0 And DE i ≤DE 0 And FH i ≤FH 0 , it is judged that the rice in the current partition is at risk of lodging;

[0091] If VC i <VC 0 And DE i >DE 0 And FH i >FH 0 , an emergency alarm is immediately issued to the current partition.

[0092] The present invention analyzes agricultural disasters in the sub-areas where the crop growth is poor. Specifically, the vegetation coverage index VC of each sub-area is calculated. i , soil environmental degradation index DE i , Pest risk index FH ik The corresponding warning values ​​VC 0 ,DE 0 , FH 0 By comparison, it is obvious that the lower the vegetation coverage rate, and the soil environment degradation index and the pest risk index have not reached the warning value, then the probability of malnutrition in the vegetation is higher, so fertilization is needed; and the higher the soil environment degradation index, the lower the vegetation coverage rate, and the pest risk index has not reached the warning value, then the probability of low vegetation coverage due to water shortage in the vegetation is higher, so irrigation is needed; and the higher the pest risk index, the lower the vegetation coverage rate, and the soil environment degradation index has not reached the warning value, then the probability of low vegetation coverage due to diseases and pests in the vegetation is higher, so disease and pest prevention and control is needed. If all three indicators reach the warning value, it means that the current zoning situation is very serious, so it is necessary to issue an emergency alarm and conduct subsequent field investigations.

[0093] The process of obtaining soil environmental degradation indicators for each sub-area is as follows:

[0094] Fitting to obtain the variation curve of soil moisture parameters over time for each partition B ω(t), and substitute it into the following formula:

[0095]

[0096] Calculate the soil environmental degradation index DE for each partition i ;

[0097] Among them, t0~t1 is the detection period, and Δθ is the reference value of the soil moisture parameter change between t0 and t1 determined based on historical data analysis.

[0098] The present invention, through the integration method Calculate the change in soil moisture parameters during the detection period. Obviously, if the cumulative change is larger, it means that there is a greater loss of soil moisture during the detection period, which means that the possibility of drought is higher, so timely irrigation is needed to maintain the stability of soil moisture.

[0099] Get the pest risk index FH for each partition i The process is:

[0100] Use the pre-trained convolutional neural network model to identify pests and diseases in standard format images, obtain the location information of each pest and disease, and then count the number of pests and diseases K1;

[0101] By formula:

[0102]

[0103] Among them, γ is the correction coefficient, which is comprehensively formulated based on historical data and empirical data, and can be adjusted according to actual conditions to consider the impact of image resolution, shooting angle, and pest size on density calculation. N is the number of sampling times, s is the sth sampling, de is the pest density, K1 is the number of pests in the current image, AZ is the current image area, is the mean density of pests and diseases.

[0104] Through the above technical solution, it is possible to Accurately calculate the density of pests and diseases in each image and then substitute it into Calculate the pest risk index FH for each partition i ; Obviously, if the volatility of pest density is greater, it means that the probability of pest infestation in the current area is higher, and therefore the risk of pest occurrence is greater.

[0105] A method for monitoring rice phenological growth based on an unmanned aerial vehicle comprises the following steps:

[0106] S1. Divide the rice field into several zones, and use the drone to fly low over the rice field according to the preset flight route and altitude, while the onboard sensor collects images of each zone;

[0107] S2, transmitting the collected images to the ground station and storing them, using image processing tools to perform pre-processing operations such as cropping, resizing, and edge cutting on the collected images to obtain images in a standard format;

[0108] S3, analyzing the preprocessed standard image, calculating the crop evaluation coefficient of each partition, and evaluating the growth of the crop according to the crop evaluation coefficient;

[0109] S4. Based on the assessment results, determine whether the rice in the current sub-area is at risk of agricultural disasters;

[0110] If there is a risk of agricultural disasters, appropriate fertilization, irrigation or pest and disease control will be carried out according to the type of agricultural disaster.

[0111] It should be noted that the calculation formulas and various parameters involved in the calculation in the present invention have been dimensionally processed in advance, and the process of dimensionless processing is well known in the industry and will not be described here.

[0112] The above is a detailed description of an embodiment of the present invention, but the content is only a preferred embodiment of the present invention and cannot be considered to limit the scope of implementation of the present invention. All equivalent changes and improvements made within the scope of the present invention should still fall within the scope of the patent coverage of the present invention.

Claims

1. A rice phenological growth monitoring system based on drone, characterized in that: include: The rice data collection module divides the rice field into several zones, and uses drones to fly low over the rice field according to a preset flight route and altitude, while the sensors carried by the drones collect images of each zone; The rice image processing module is used to transmit the collected images to the ground station and store them, and use image processing tools to perform pre-processing operations such as cropping, resizing, and edge cutting on the collected images to obtain images in a standard format; The rice data analysis module is used to analyze the pre-processed standard image and calculate the crop evaluation coefficient of each partition, and evaluate the growth of the crop according to the crop evaluation coefficient; The response strategy module determines whether the rice in the current partition is at risk of agricultural disasters based on the assessment results; if there is a risk of agricultural disasters, corresponding fertilization, irrigation or pest control is carried out according to the type of agricultural disaster, which includes lodging, drought and insect pests.

2. The rice phenological growth monitoring system based on drone according to claim 1 is characterized in that: The process of obtaining the crop evaluation coefficient for each partition is: By formula: Calculate the crop assessment coefficient Q for each partition i ; Among them, i is the i-th partition, VC i is the vegetation coverage index for each partition, α i is the level of rice growth potential in each partition; when the rice growth potential level is low, α i =1, when the rice growth potential level is medium, α i =2, when the rice growth potential level is high, α i =3; VC ith is the standard value of vegetation coverage index, DE th DE is the reference value of soil environmental degradation index. i is the soil environmental degradation index for each sub-area, FH ik For each sub-area, the pest risk index is is the average value of the pest risk index, and μ and ρ are adjustment coefficients.

3. The rice phenological growth monitoring system based on drone according to claim 2 is characterized in that: The process of determining the rice growth potential level in each partition is as follows: The processed image was binarized using the adaptive threshold method Otsu to obtain a binary image, in which white pixels represent the rice area and black pixels represent the background area; According to the pixel value R of the white area in the binary image j Calculate the vegetation coverage index VC for each partition i , the expression is: Among them, W i is the total pixel value, n is the number of images, and j is the jth image; The vegetation coverage index VC of each partition i With the preset preparation coverage threshold interval [VC min , VC max ] for comparison; If VC i >VC max , then the rice growth potential is judged to be advanced; If VC i ∈[VC min , VC max ], the rice growth potential is judged to be medium; If VC i <VC min , then the level of rice growth potential is judged to be low.

4. The rice phenological growth monitoring system based on drone according to claim 3 is characterized in that: The process of evaluating the growth of crops according to the crop evaluation coefficient is as follows: The crop evaluation coefficient Q for each partition obtained will be calculated i Compare with the preset crop assessment threshold Q0; If Q i ≥Q0, it is estimated that the crop growth in the current partition is good; If Q i <is less than Q0, the preliminary assessment is that the crop growth in the current partition is poor.

5. The rice phenological growth monitoring system based on drone according to claim 4 is characterized in that: According to the assessment results, the process of judging whether there is a risk of agricultural disasters in the current sub-area is as follows: A secondary judgment is made on the zones where the crop growth is poor. The specific process is as follows: The vegetation coverage index VC of each partition i , soil environmental degradation index DE i , Pest risk index FH ik Compare with the corresponding warning values ​​VC0, DE0, and FH0 respectively; If VC i <VC0 and DE i >DE0 and FH i ≤FH0, then it is determined that there is a drought risk for the rice in the current partition; If VC i <VC0 and DE i ≤DE0 and FH i >FH0, then it is determined that there is a risk of pests and diseases in the rice in the current partition; If VC i ≥VC0 and DE i ≤DE0 and FH i ≤FH0, it is judged that the rice in the current partition is at risk of lodging; If VC i <VC < 0 and DE i > DE0 and FH i > FH0, an emergency alarm is immediately issued for the current partition.

6. The rice phenological growth monitoring system based on drone according to claim 2 is characterized in that: The process of obtaining soil environmental degradation indicators for each sub-area is as follows: Fitting to obtain the variation curve of soil moisture parameters over time for each partition B ω (t), and substitute it into the following formula: Calculate the soil environmental degradation index DE for each partition i ; Among them, t0~t1 is the detection period, and Δθ is the reference value of the soil moisture parameter change between t0 and t1 determined based on historical data analysis.

7. The rice phenological growth monitoring system based on drone according to claim 2 is characterized in that: Get the pest risk index FH for each partition i The process is: Use the pre-trained convolutional neural network model to identify pests and diseases in standard format images, obtain the location information of each pest and disease, and then count the number of pests and diseases K1; By formula: Among them, γ is the correction coefficient, N is the number of sampling times, s is the sth sampling, de is the density of pests and diseases, K1 is the number of pests and diseases in the current image, AZ is the area of ​​the current image, is the mean density of pests and diseases.

8. A method for monitoring rice phenological growth based on drones, the method being implemented based on the rice phenological growth monitoring system based on drones as claimed in any one of claims 1 to 7, characterized in that: The steps include: S1. Divide the rice field into several zones, and use the drone to fly low over the rice field according to the preset flight route and altitude, while the onboard sensor collects images of each zone; S2, transmitting the collected images to the ground station and storing them, using image processing tools to perform pre-processing operations such as cropping, resizing, and edge cutting on the collected images to obtain images in a standard format; S3, analyzing the preprocessed standard image, calculating the crop evaluation coefficient of each partition, and evaluating the growth of the crop according to the crop evaluation coefficient; S4. Based on the assessment results, determine whether the rice in the current sub-area is at risk of agricultural disasters; If there is a risk of agricultural disasters, appropriate fertilization, irrigation or pest and disease control will be carried out according to the type of agricultural disaster.

Images